Fix a (rare) race condition in mv_interrupt() when using MSI.
The value of hpriv->main_irq_mask_addr can change on on the fly,
and without this patch we could end up writing back a stale copy
to the hardware.
Signed-off-by: Mark Lord <mlord@pobox.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
else
handled = mv_host_intr(host, pending_irqs);
}
- spin_unlock(&host->lock);
/* for MSI: unmask; interrupt cause bits will retrigger now */
if (using_msi)
writel(hpriv->main_irq_mask, hpriv->main_irq_mask_addr);
+ spin_unlock(&host->lock);
+
return IRQ_RETVAL(handled);
}